ripemd.cpp 33.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
/* ripemd.cpp                                
 *
 * Copyright (C) 2003 Sawtooth Consulting Ltd.
 *
 * This file is part of yaSSL.
 *
 * yaSSL is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 2 of the License, or
 * (at your option) any later version.
 *
 * yaSSL is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
 */


/* based on Wei Dai's ripemd.cpp from CryptoPP */

#include "runtime.hpp"
#include "ripemd.hpp"
#include "algorithm.hpp"    // mySTL::swap

unknown's avatar
unknown committed
29 30 31 32 33 34


#if defined(TAOCRYPT_X86ASM_AVAILABLE) && defined(TAO_ASM)
    #define DO_RIPEMD_ASM
#endif

35 36 37 38 39 40 41 42 43 44 45
namespace TaoCrypt {

void RIPEMD160::Init()
{
    digest_[0] = 0x67452301L;
    digest_[1] = 0xefcdab89L;
    digest_[2] = 0x98badcfeL;
    digest_[3] = 0x10325476L;
    digest_[4] = 0xc3d2e1f0L;

    buffLen_ = 0;
unknown's avatar
unknown committed
46 47
    loLen_  = 0;
    hiLen_  = 0;
48 49 50 51 52 53 54
}


RIPEMD160::RIPEMD160(const RIPEMD160& that)
    : HASHwithTransform(DIGEST_SIZE / sizeof(word32), BLOCK_SIZE) 
{ 
    buffLen_ = that.buffLen_;
unknown's avatar
unknown committed
55 56
    loLen_   = that.loLen_;
    hiLen_   = that.hiLen_;
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73

    memcpy(digest_, that.digest_, DIGEST_SIZE);
    memcpy(buffer_, that.buffer_, BLOCK_SIZE);
}


RIPEMD160& RIPEMD160::operator= (const RIPEMD160& that)
{
    RIPEMD160 tmp(that);
    Swap(tmp);

    return *this;
}


void RIPEMD160::Swap(RIPEMD160& other)
{
unknown's avatar
unknown committed
74 75
    mySTL::swap(loLen_,   other.loLen_);
    mySTL::swap(hiLen_,   other.hiLen_);
unknown's avatar
unknown committed
76 77 78 79
    mySTL::swap(buffLen_, other.buffLen_);

    memcpy(digest_, other.digest_, DIGEST_SIZE);
    memcpy(buffer_, other.buffer_, BLOCK_SIZE);
80 81 82
}


unknown's avatar
unknown committed
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137
// Update digest with data of size len, do in blocks
void RIPEMD160::Update(const byte* data, word32 len)
{
    byte* local = (byte*)buffer_;

    // remove buffered data if possible
    if (buffLen_)  {   
        word32 add = min(len, BLOCK_SIZE - buffLen_);
        memcpy(&local[buffLen_], data, add);

        buffLen_ += add;
        data     += add;
        len      -= add;

        if (buffLen_ == BLOCK_SIZE) {
            ByteReverseIf(local, local, BLOCK_SIZE, LittleEndianOrder);
            Transform();
            AddLength(BLOCK_SIZE);
            buffLen_ = 0;
        }
    }

    // do block size transforms or all at once for asm
    if (buffLen_ == 0) {
        #ifndef DO_RIPEMD_ASM
            while (len >= BLOCK_SIZE) {
                memcpy(&local[0], data, BLOCK_SIZE);

                data     += BLOCK_SIZE;
                len      -= BLOCK_SIZE;

                ByteReverseIf(local, local, BLOCK_SIZE, LittleEndianOrder);
                Transform();
                AddLength(BLOCK_SIZE);
            }
        #else
            word32 times = len / BLOCK_SIZE;
            if (times) {
                AsmTransform(data, times);
                const word32 add = BLOCK_SIZE * times;
                AddLength(add);
                len  -= add;
                data += add;
            }
        #endif
    }

    // cache any data left
    if (len) {
        memcpy(&local[buffLen_], data, len);
        buffLen_ += len;
    }
}


138 139 140 141 142 143 144 145
// for all
#define F(x, y, z)    (x ^ y ^ z) 
#define G(x, y, z)    (z ^ (x & (y^z)))
#define H(x, y, z)    (z ^ (x | ~y))
#define I(x, y, z)    (y ^ (z & (x^y)))
#define J(x, y, z)    (x ^ (y | ~z))

#define k0 0
unknown's avatar
unknown committed
146 147 148 149 150 151 152 153
#define k1 0x5a827999
#define k2 0x6ed9eba1
#define k3 0x8f1bbcdc
#define k4 0xa953fd4e
#define k5 0x50a28be6
#define k6 0x5c4dd124
#define k7 0x6d703ef3
#define k8 0x7a6d76e9
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171
#define k9 0

// for 160 and 320
#define Subround(f, a, b, c, d, e, x, s, k) \
    a += f(b, c, d) + x + k;\
    a = rotlFixed((word32)a, s) + e;\
    c = rotlFixed((word32)c, 10U)


void RIPEMD160::Transform()
{
    unsigned long a1, b1, c1, d1, e1, a2, b2, c2, d2, e2;
    a1 = a2 = digest_[0];
    b1 = b2 = digest_[1];
    c1 = c2 = digest_[2];
    d1 = d2 = digest_[3];
    e1 = e2 = digest_[4];

unknown's avatar
unknown committed
172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
    Subround(F, a1, b1, c1, d1, e1, buffer_[ 0], 11, k0);
    Subround(F, e1, a1, b1, c1, d1, buffer_[ 1], 14, k0);
    Subround(F, d1, e1, a1, b1, c1, buffer_[ 2], 15, k0);
    Subround(F, c1, d1, e1, a1, b1, buffer_[ 3], 12, k0);
    Subround(F, b1, c1, d1, e1, a1, buffer_[ 4],  5, k0);
    Subround(F, a1, b1, c1, d1, e1, buffer_[ 5],  8, k0);
    Subround(F, e1, a1, b1, c1, d1, buffer_[ 6],  7, k0);
    Subround(F, d1, e1, a1, b1, c1, buffer_[ 7],  9, k0);
    Subround(F, c1, d1, e1, a1, b1, buffer_[ 8], 11, k0);
    Subround(F, b1, c1, d1, e1, a1, buffer_[ 9], 13, k0);
    Subround(F, a1, b1, c1, d1, e1, buffer_[10], 14, k0);
    Subround(F, e1, a1, b1, c1, d1, buffer_[11], 15, k0);
    Subround(F, d1, e1, a1, b1, c1, buffer_[12],  6, k0);
    Subround(F, c1, d1, e1, a1, b1, buffer_[13],  7, k0);
    Subround(F, b1, c1, d1, e1, a1, buffer_[14],  9, k0);
    Subround(F, a1, b1, c1, d1, e1, buffer_[15],  8, k0);

    Subround(G, e1, a1, b1, c1, d1, buffer_[ 7],  7, k1);
    Subround(G, d1, e1, a1, b1, c1, buffer_[ 4],  6, k1);
    Subround(G, c1, d1, e1, a1, b1, buffer_[13],  8, k1);
    Subround(G, b1, c1, d1, e1, a1, buffer_[ 1], 13, k1);
    Subround(G, a1, b1, c1, d1, e1, buffer_[10], 11, k1);
    Subround(G, e1, a1, b1, c1, d1, buffer_[ 6],  9, k1);
    Subround(G, d1, e1, a1, b1, c1, buffer_[15],  7, k1);
    Subround(G, c1, d1, e1, a1, b1, buffer_[ 3], 15, k1);
    Subround(G, b1, c1, d1, e1, a1, buffer_[12],  7, k1);
    Subround(G, a1, b1, c1, d1, e1, buffer_[ 0], 12, k1);
    Subround(G, e1, a1, b1, c1, d1, buffer_[ 9], 15, k1);
    Subround(G, d1, e1, a1, b1, c1, buffer_[ 5],  9, k1);
    Subround(G, c1, d1, e1, a1, b1, buffer_[ 2], 11, k1);
    Subround(G, b1, c1, d1, e1, a1, buffer_[14],  7, k1);
    Subround(G, a1, b1, c1, d1, e1, buffer_[11], 13, k1);
    Subround(G, e1, a1, b1, c1, d1, buffer_[ 8], 12, k1);

    Subround(H, d1, e1, a1, b1, c1, buffer_[ 3], 11, k2);
    Subround(H, c1, d1, e1, a1, b1, buffer_[10], 13, k2);
    Subround(H, b1, c1, d1, e1, a1, buffer_[14],  6, k2);
    Subround(H, a1, b1, c1, d1, e1, buffer_[ 4],  7, k2);
    Subround(H, e1, a1, b1, c1, d1, buffer_[ 9], 14, k2);
    Subround(H, d1, e1, a1, b1, c1, buffer_[15],  9, k2);
    Subround(H, c1, d1, e1, a1, b1, buffer_[ 8], 13, k2);
    Subround(H, b1, c1, d1, e1, a1, buffer_[ 1], 15, k2);
    Subround(H, a1, b1, c1, d1, e1, buffer_[ 2], 14, k2);
    Subround(H, e1, a1, b1, c1, d1, buffer_[ 7],  8, k2);
    Subround(H, d1, e1, a1, b1, c1, buffer_[ 0], 13, k2);
    Subround(H, c1, d1, e1, a1, b1, buffer_[ 6],  6, k2);
    Subround(H, b1, c1, d1, e1, a1, buffer_[13],  5, k2);
    Subround(H, a1, b1, c1, d1, e1, buffer_[11], 12, k2);
    Subround(H, e1, a1, b1, c1, d1, buffer_[ 5],  7, k2);
    Subround(H, d1, e1, a1, b1, c1, buffer_[12],  5, k2);

    Subround(I, c1, d1, e1, a1, b1, buffer_[ 1], 11, k3);
    Subround(I, b1, c1, d1, e1, a1, buffer_[ 9], 12, k3);
    Subround(I, a1, b1, c1, d1, e1, buffer_[11], 14, k3);
    Subround(I, e1, a1, b1, c1, d1, buffer_[10], 15, k3);
    Subround(I, d1, e1, a1, b1, c1, buffer_[ 0], 14, k3);
    Subround(I, c1, d1, e1, a1, b1, buffer_[ 8], 15, k3);
    Subround(I, b1, c1, d1, e1, a1, buffer_[12],  9, k3);
    Subround(I, a1, b1, c1, d1, e1, buffer_[ 4],  8, k3);
    Subround(I, e1, a1, b1, c1, d1, buffer_[13],  9, k3);
    Subround(I, d1, e1, a1, b1, c1, buffer_[ 3], 14, k3);
    Subround(I, c1, d1, e1, a1, b1, buffer_[ 7],  5, k3);
    Subround(I, b1, c1, d1, e1, a1, buffer_[15],  6, k3);
    Subround(I, a1, b1, c1, d1, e1, buffer_[14],  8, k3);
    Subround(I, e1, a1, b1, c1, d1, buffer_[ 5],  6, k3);
    Subround(I, d1, e1, a1, b1, c1, buffer_[ 6],  5, k3);
    Subround(I, c1, d1, e1, a1, b1, buffer_[ 2], 12, k3);

    Subround(J, b1, c1, d1, e1, a1, buffer_[ 4],  9, k4);
    Subround(J, a1, b1, c1, d1, e1, buffer_[ 0], 15, k4);
    Subround(J, e1, a1, b1, c1, d1, buffer_[ 5],  5, k4);
    Subround(J, d1, e1, a1, b1, c1, buffer_[ 9], 11, k4);
    Subround(J, c1, d1, e1, a1, b1, buffer_[ 7],  6, k4);
    Subround(J, b1, c1, d1, e1, a1, buffer_[12],  8, k4);
    Subround(J, a1, b1, c1, d1, e1, buffer_[ 2], 13, k4);
    Subround(J, e1, a1, b1, c1, d1, buffer_[10], 12, k4);
    Subround(J, d1, e1, a1, b1, c1, buffer_[14],  5, k4);
    Subround(J, c1, d1, e1, a1, b1, buffer_[ 1], 12, k4);
    Subround(J, b1, c1, d1, e1, a1, buffer_[ 3], 13, k4);
    Subround(J, a1, b1, c1, d1, e1, buffer_[ 8], 14, k4);
    Subround(J, e1, a1, b1, c1, d1, buffer_[11], 11, k4);
    Subround(J, d1, e1, a1, b1, c1, buffer_[ 6],  8, k4);
    Subround(J, c1, d1, e1, a1, b1, buffer_[15],  5, k4);
    Subround(J, b1, c1, d1, e1, a1, buffer_[13],  6, k4);

    Subround(J, a2, b2, c2, d2, e2, buffer_[ 5],  8, k5);
    Subround(J, e2, a2, b2, c2, d2, buffer_[14],  9, k5);
    Subround(J, d2, e2, a2, b2, c2, buffer_[ 7],  9, k5);
    Subround(J, c2, d2, e2, a2, b2, buffer_[ 0], 11, k5);
    Subround(J, b2, c2, d2, e2, a2, buffer_[ 9], 13, k5);
    Subround(J, a2, b2, c2, d2, e2, buffer_[ 2], 15, k5);
    Subround(J, e2, a2, b2, c2, d2, buffer_[11], 15, k5);
    Subround(J, d2, e2, a2, b2, c2, buffer_[ 4],  5, k5);
    Subround(J, c2, d2, e2, a2, b2, buffer_[13],  7, k5);
    Subround(J, b2, c2, d2, e2, a2, buffer_[ 6],  7, k5);
    Subround(J, a2, b2, c2, d2, e2, buffer_[15],  8, k5);
    Subround(J, e2, a2, b2, c2, d2, buffer_[ 8], 11, k5);
    Subround(J, d2, e2, a2, b2, c2, buffer_[ 1], 14, k5);
    Subround(J, c2, d2, e2, a2, b2, buffer_[10], 14, k5);
    Subround(J, b2, c2, d2, e2, a2, buffer_[ 3], 12, k5);
    Subround(J, a2, b2, c2, d2, e2, buffer_[12],  6, k5);

    Subround(I, e2, a2, b2, c2, d2, buffer_[ 6],  9, k6); 
    Subround(I, d2, e2, a2, b2, c2, buffer_[11], 13, k6);
    Subround(I, c2, d2, e2, a2, b2, buffer_[ 3], 15, k6);
    Subround(I, b2, c2, d2, e2, a2, buffer_[ 7],  7, k6);
    Subround(I, a2, b2, c2, d2, e2, buffer_[ 0], 12, k6);
    Subround(I, e2, a2, b2, c2, d2, buffer_[13],  8, k6);
    Subround(I, d2, e2, a2, b2, c2, buffer_[ 5],  9, k6);
    Subround(I, c2, d2, e2, a2, b2, buffer_[10], 11, k6);
    Subround(I, b2, c2, d2, e2, a2, buffer_[14],  7, k6);
    Subround(I, a2, b2, c2, d2, e2, buffer_[15],  7, k6);
    Subround(I, e2, a2, b2, c2, d2, buffer_[ 8], 12, k6);
    Subround(I, d2, e2, a2, b2, c2, buffer_[12],  7, k6);
    Subround(I, c2, d2, e2, a2, b2, buffer_[ 4],  6, k6);
    Subround(I, b2, c2, d2, e2, a2, buffer_[ 9], 15, k6);
    Subround(I, a2, b2, c2, d2, e2, buffer_[ 1], 13, k6);
    Subround(I, e2, a2, b2, c2, d2, buffer_[ 2], 11, k6);

    Subround(H, d2, e2, a2, b2, c2, buffer_[15],  9, k7);
    Subround(H, c2, d2, e2, a2, b2, buffer_[ 5],  7, k7);
    Subround(H, b2, c2, d2, e2, a2, buffer_[ 1], 15, k7);
    Subround(H, a2, b2, c2, d2, e2, buffer_[ 3], 11, k7);
    Subround(H, e2, a2, b2, c2, d2, buffer_[ 7],  8, k7);
    Subround(H, d2, e2, a2, b2, c2, buffer_[14],  6, k7);
    Subround(H, c2, d2, e2, a2, b2, buffer_[ 6],  6, k7);
    Subround(H, b2, c2, d2, e2, a2, buffer_[ 9], 14, k7);
    Subround(H, a2, b2, c2, d2, e2, buffer_[11], 12, k7);
    Subround(H, e2, a2, b2, c2, d2, buffer_[ 8], 13, k7);
    Subround(H, d2, e2, a2, b2, c2, buffer_[12],  5, k7);
    Subround(H, c2, d2, e2, a2, b2, buffer_[ 2], 14, k7);
    Subround(H, b2, c2, d2, e2, a2, buffer_[10], 13, k7);
    Subround(H, a2, b2, c2, d2, e2, buffer_[ 0], 13, k7);
    Subround(H, e2, a2, b2, c2, d2, buffer_[ 4],  7, k7);
    Subround(H, d2, e2, a2, b2, c2, buffer_[13],  5, k7);

    Subround(G, c2, d2, e2, a2, b2, buffer_[ 8], 15, k8);
    Subround(G, b2, c2, d2, e2, a2, buffer_[ 6],  5, k8);
    Subround(G, a2, b2, c2, d2, e2, buffer_[ 4],  8, k8);
    Subround(G, e2, a2, b2, c2, d2, buffer_[ 1], 11, k8);
    Subround(G, d2, e2, a2, b2, c2, buffer_[ 3], 14, k8);
    Subround(G, c2, d2, e2, a2, b2, buffer_[11], 14, k8);
    Subround(G, b2, c2, d2, e2, a2, buffer_[15],  6, k8);
    Subround(G, a2, b2, c2, d2, e2, buffer_[ 0], 14, k8);
    Subround(G, e2, a2, b2, c2, d2, buffer_[ 5],  6, k8);
    Subround(G, d2, e2, a2, b2, c2, buffer_[12],  9, k8);
    Subround(G, c2, d2, e2, a2, b2, buffer_[ 2], 12, k8);
    Subround(G, b2, c2, d2, e2, a2, buffer_[13],  9, k8);
    Subround(G, a2, b2, c2, d2, e2, buffer_[ 9], 12, k8);
    Subround(G, e2, a2, b2, c2, d2, buffer_[ 7],  5, k8);
    Subround(G, d2, e2, a2, b2, c2, buffer_[10], 15, k8);
    Subround(G, c2, d2, e2, a2, b2, buffer_[14],  8, k8);

    Subround(F, b2, c2, d2, e2, a2, buffer_[12],  8, k9);
    Subround(F, a2, b2, c2, d2, e2, buffer_[15],  5, k9);
    Subround(F, e2, a2, b2, c2, d2, buffer_[10], 12, k9);
    Subround(F, d2, e2, a2, b2, c2, buffer_[ 4],  9, k9);
    Subround(F, c2, d2, e2, a2, b2, buffer_[ 1], 12, k9);
    Subround(F, b2, c2, d2, e2, a2, buffer_[ 5],  5, k9);
    Subround(F, a2, b2, c2, d2, e2, buffer_[ 8], 14, k9);
    Subround(F, e2, a2, b2, c2, d2, buffer_[ 7],  6, k9);
    Subround(F, d2, e2, a2, b2, c2, buffer_[ 6],  8, k9);
    Subround(F, c2, d2, e2, a2, b2, buffer_[ 2], 13, k9);
    Subround(F, b2, c2, d2, e2, a2, buffer_[13],  6, k9);
    Subround(F, a2, b2, c2, d2, e2, buffer_[14],  5, k9);
    Subround(F, e2, a2, b2, c2, d2, buffer_[ 0], 15, k9);
    Subround(F, d2, e2, a2, b2, c2, buffer_[ 3], 13, k9);
    Subround(F, c2, d2, e2, a2, b2, buffer_[ 9], 11, k9);
    Subround(F, b2, c2, d2, e2, a2, buffer_[11], 11, k9);
341 342 343 344 345 346 347

    c1         = digest_[1] + c1 + d2;
    digest_[1] = digest_[2] + d1 + e2;
    digest_[2] = digest_[3] + e1 + a2;
    digest_[3] = digest_[4] + a1 + b2;
    digest_[4] = digest_[0] + b1 + c2;
    digest_[0] = c1;
unknown's avatar
unknown committed
348
}
349

unknown's avatar
unknown committed
350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832

#ifdef DO_RIPEMD_ASM

/*
    // F(x ^ y ^ z)
    // place in esi
#define ASMF(x, y, z)  \
    AS2(    mov   esi, x                )   \
    AS2(    xor   esi, y                )   \
    AS2(    xor   esi, z                )


    // G(z ^ (x & (y^z)))
    // place in esi
#define ASMG(x, y, z)  \
    AS2(    mov   esi, z                )   \
    AS2(    xor   esi, y                )   \
    AS2(    and   esi, x                )   \
    AS2(    xor   esi, z                )

    
    // H(z ^ (x | ~y))
    // place in esi
#define ASMH(x, y, z) \
    AS2(    mov   esi, y                )   \
    AS1(    not   esi                   )   \
    AS2(     or   esi, x                )   \
    AS2(    xor   esi, z                )


    // I(y ^ (z & (x^y)))
    // place in esi
#define ASMI(x, y, z)  \
    AS2(    mov   esi, y                )   \
    AS2(    xor   esi, x                )   \
    AS2(    and   esi, z                )   \
    AS2(    xor   esi, y                )


    // J(x ^ (y | ~z)))
    // place in esi
#define ASMJ(x, y, z)   \
    AS2(    mov   esi, z                )   \
    AS1(    not   esi                   )   \
    AS2(     or   esi, y                )   \
    AS2(    xor   esi, x                )


// for 160 and 320
// #define ASMSubround(f, a, b, c, d, e, i, s, k) 
//    a += f(b, c, d) + data[i] + k;
//    a = rotlFixed((word32)a, s) + e;
//    c = rotlFixed((word32)c, 10U)

#define ASMSubround(f, a, b, c, d, e, index, s, k) \
    // a += f(b, c, d) + data[i] + k                    \
    AS2(    mov   esp, [edi + index * 4]            )   \
    f(b, c, d)                                          \
    AS2(    add   esi, k                            )   \
    AS2(    add   esi, esp                          )   \
    AS2(    add     a, esi                          )   \
    // a = rotlFixed((word32)a, s) + e                  \
    AS2(    rol     a, s                            )   \
    AS2(    rol     c, 10                           )   \
    // c = rotlFixed((word32)c, 10U)                    \
    AS2(    add     a, e                            )
*/


// combine F into subround w/ setup
// esi already has c, setup for next round when done
// esp already has edi[index], setup for next round when done

#define ASMSubroundF(a, b, c, d, e, index, s) \
    /* a += (b ^ c ^ d) + data[i] + k  */               \
    AS2(    xor   esi, b                            )   \
    AS2(    add     a, [edi + index * 4]            )   \
    AS2(    xor   esi, d                            )   \
    AS2(    add     a, esi                          )   \
    /* a = rotlFixed((word32)a, s) + e */               \
    AS2(    mov   esi, b                            )   \
    AS2(    rol     a, s                            )   \
    /* c = rotlFixed((word32)c, 10U) */                 \
    AS2(    rol     c, 10                           )   \
    AS2(    add     a, e                            )


// combine G into subround w/ setup
// esi already has c, setup for next round when done
// esp already has edi[index], setup for next round when done

#define ASMSubroundG(a, b, c, d, e, index, s, k) \
    /* a += (d ^ (b & (c^d))) + data[i] + k  */         \
    AS2(    xor   esi, d                            )   \
    AS2(    and   esi, b                            )   \
    AS2(    add     a, [edi + index * 4]            )   \
    AS2(    xor   esi, d                            )   \
    AS2(    lea     a, [esi + a + k]                )   \
    /* a = rotlFixed((word32)a, s) + e */               \
    AS2(    mov   esi, b                            )   \
    AS2(    rol     a, s                            )   \
    /* c = rotlFixed((word32)c, 10U) */                 \
    AS2(    rol     c, 10                           )   \
    AS2(    add     a, e                            )


// combine H into subround w/ setup
// esi already has c, setup for next round when done
// esp already has edi[index], setup for next round when done

#define ASMSubroundH(a, b, c, d, e, index, s, k) \
    /* a += (d ^ (b | ~c)) + data[i] + k  */            \
    AS1(    not   esi                               )   \
    AS2(     or   esi, b                            )   \
    AS2(    add     a, [edi + index * 4]            )   \
    AS2(    xor   esi, d                            )   \
    AS2(    lea     a, [esi + a + k]                )   \
    /* a = rotlFixed((word32)a, s) + e */               \
    AS2(    mov   esi, b                            )   \
    AS2(    rol     a, s                            )   \
    /* c = rotlFixed((word32)c, 10U) */                 \
    AS2(    rol     c, 10                           )   \
    AS2(    add     a, e                            )


// combine I into subround w/ setup
// esi already has c, setup for next round when done
// esp already has edi[index], setup for next round when done

#define ASMSubroundI(a, b, c, d, e, index, s, k) \
    /* a += (c ^ (d & (b^c))) + data[i] + k  */         \
    AS2(    xor   esi, b                            )   \
    AS2(    and   esi, d                            )   \
    AS2(    add     a, [edi + index * 4]            )   \
    AS2(    xor   esi, c                            )   \
    AS2(    lea     a, [esi + a + k]                )   \
    /* a = rotlFixed((word32)a, s) + e */               \
    AS2(    mov   esi, b                            )   \
    AS2(    rol     a, s                            )   \
    /* c = rotlFixed((word32)c, 10U) */                 \
    AS2(    rol     c, 10                           )   \
    AS2(    add     a, e                            )


// combine J into subround w/ setup
// esi already has d, setup for next round when done
// esp already has edi[index], setup for next round when done

#define ASMSubroundJ(a, b, c, d, e, index, s, k) \
    /* a += (b ^ (c | ~d))) + data[i] + k  */           \
    AS1(    not   esi                               )   \
    AS2(     or   esi, c                            )   \
    /* c = rotlFixed((word32)c, 10U) */                 \
    AS2(    add     a, [edi + index * 4]            )   \
    AS2(    xor   esi, b                            )   \
    AS2(    rol     c, 10                           )   \
    AS2(    lea     a, [esi + a + k]                )   \
    /* a = rotlFixed((word32)a, s) + e */               \
    AS2(    rol     a, s                            )   \
    AS2(    mov   esi, c                            )   \
    AS2(    add     a, e                            )


#ifdef _MSC_VER
    __declspec(naked) 
#endif
void RIPEMD160::AsmTransform(const byte* data, word32 times)
{
#ifdef __GNUC__
    #define AS1(x)    asm(#x);
    #define AS2(x, y) asm(#x ", " #y);

    #define PROLOG()  \
        asm(".intel_syntax noprefix"); \
        AS2(    movd  mm3, edi                      )   \
        AS2(    movd  mm4, ebx                      )   \
        AS2(    movd  mm5, esi                      )   \
        AS2(    movd  mm6, ebp                      )   \
        AS2(    mov   ecx, DWORD PTR [ebp +  8]     )   \
        AS2(    mov   edi, DWORD PTR [ebp + 12]     )   \
        AS2(    mov   edx, DWORD PTR [ebp + 16]     )

    #define EPILOG()  \
        AS2(    movd  ebp, mm6                  )   \
        AS2(    movd  esi, mm5                  )   \
        AS2(    movd  ebx, mm4                  )   \
        AS2(    mov   esp, ebp                  )   \
        AS2(    movd  edi, mm3                  )   \
        AS1(    emms                            )   \
        asm(".att_syntax");
#else
    #define AS1(x)    __asm x
    #define AS2(x, y) __asm x, y

    #define PROLOG() \
        AS1(    push  ebp                       )   \
        AS2(    mov   ebp, esp                  )   \
        AS2(    movd  mm3, edi                  )   \
        AS2(    movd  mm4, ebx                  )   \
        AS2(    movd  mm5, esi                  )   \
        AS2(    movd  mm6, ebp                  )   \
        AS2(    mov   edi, DWORD PTR [ebp +  8] )   \
        AS2(    mov   edx, DWORD PTR [ebp + 12] )

    #define EPILOG() \
        AS2(    movd  ebp, mm6                  )   \
        AS2(    movd  esi, mm5                  )   \
        AS2(    movd  ebx, mm4                  )   \
        AS2(    movd  edi, mm3                  )   \
        AS2(    mov   esp, ebp                  )   \
        AS1(    pop   ebp                       )   \
        AS1(    emms                            )   \
        AS1(    ret   8                         )
        
#endif

    PROLOG()

    #ifdef OLD_GCC_OFFSET
        AS2(    lea   esi, [ecx + 20]               )   // digest_[0]
    #else
        AS2(    lea   esi, [ecx + 16]               )   // digest_[0]
    #endif

    AS2(    sub   esp, 24               )   // make room for tmp a1 - e1
    AS2(    movd  mm1, esi              )   // store digest_
    
AS1( loopStart: )

    AS2(    movd  mm2, edx              )   // store times_

    AS2(    mov   eax, [esi]            )   // a1
    AS2(    mov   ebx, [esi +  4]       )   // b1
    AS2(    mov   ecx, [esi +  8]       )   // c1
    AS2(    mov   edx, [esi + 12]       )   // d1
    AS2(    mov   ebp, [esi + 16]       )   // e1

    // setup 
    AS2(    mov   esi, ecx      )

    ASMSubroundF( eax, ebx, ecx, edx, ebp,  0, 11)
    ASMSubroundF( ebp, eax, ebx, ecx, edx,  1, 14)
    ASMSubroundF( edx, ebp, eax, ebx, ecx,  2, 15)
    ASMSubroundF( ecx, edx, ebp, eax, ebx,  3, 12)
    ASMSubroundF( ebx, ecx, edx, ebp, eax,  4,  5)
    ASMSubroundF( eax, ebx, ecx, edx, ebp,  5,  8)
    ASMSubroundF( ebp, eax, ebx, ecx, edx,  6,  7)
    ASMSubroundF( edx, ebp, eax, ebx, ecx,  7,  9)
    ASMSubroundF( ecx, edx, ebp, eax, ebx,  8, 11)
    ASMSubroundF( ebx, ecx, edx, ebp, eax,  9, 13)
    ASMSubroundF( eax, ebx, ecx, edx, ebp, 10, 14)
    ASMSubroundF( ebp, eax, ebx, ecx, edx, 11, 15)
    ASMSubroundF( edx, ebp, eax, ebx, ecx, 12,  6)
    ASMSubroundF( ecx, edx, ebp, eax, ebx, 13,  7)
    ASMSubroundF( ebx, ecx, edx, ebp, eax, 14,  9)
    ASMSubroundF( eax, ebx, ecx, edx, ebp, 15,  8)

    ASMSubroundG( ebp, eax, ebx, ecx, edx,  7,  7, k1)
    ASMSubroundG( edx, ebp, eax, ebx, ecx,  4,  6, k1)
    ASMSubroundG( ecx, edx, ebp, eax, ebx, 13,  8, k1)
    ASMSubroundG( ebx, ecx, edx, ebp, eax,  1, 13, k1)
    ASMSubroundG( eax, ebx, ecx, edx, ebp, 10, 11, k1)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  6,  9, k1)
    ASMSubroundG( edx, ebp, eax, ebx, ecx, 15,  7, k1)
    ASMSubroundG( ecx, edx, ebp, eax, ebx,  3, 15, k1)
    ASMSubroundG( ebx, ecx, edx, ebp, eax, 12,  7, k1)
    ASMSubroundG( eax, ebx, ecx, edx, ebp,  0, 12, k1)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  9, 15, k1)
    ASMSubroundG( edx, ebp, eax, ebx, ecx,  5,  9, k1)
    ASMSubroundG( ecx, edx, ebp, eax, ebx,  2, 11, k1)
    ASMSubroundG( ebx, ecx, edx, ebp, eax, 14,  7, k1)
    ASMSubroundG( eax, ebx, ecx, edx, ebp, 11, 13, k1)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  8, 12, k1)

    ASMSubroundH( edx, ebp, eax, ebx, ecx,  3, 11, k2)
    ASMSubroundH( ecx, edx, ebp, eax, ebx, 10, 13, k2)
    ASMSubroundH( ebx, ecx, edx, ebp, eax, 14,  6, k2)
    ASMSubroundH( eax, ebx, ecx, edx, ebp,  4,  7, k2)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  9, 14, k2)
    ASMSubroundH( edx, ebp, eax, ebx, ecx, 15,  9, k2)
    ASMSubroundH( ecx, edx, ebp, eax, ebx,  8, 13, k2)
    ASMSubroundH( ebx, ecx, edx, ebp, eax,  1, 15, k2)
    ASMSubroundH( eax, ebx, ecx, edx, ebp,  2, 14, k2)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  7,  8, k2)
    ASMSubroundH( edx, ebp, eax, ebx, ecx,  0, 13, k2)
    ASMSubroundH( ecx, edx, ebp, eax, ebx,  6,  6, k2)
    ASMSubroundH( ebx, ecx, edx, ebp, eax, 13,  5, k2)
    ASMSubroundH( eax, ebx, ecx, edx, ebp, 11, 12, k2)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  5,  7, k2)
    ASMSubroundH( edx, ebp, eax, ebx, ecx, 12,  5, k2)

    ASMSubroundI( ecx, edx, ebp, eax, ebx,  1, 11, k3)
    ASMSubroundI( ebx, ecx, edx, ebp, eax,  9, 12, k3)
    ASMSubroundI( eax, ebx, ecx, edx, ebp, 11, 14, k3)
    ASMSubroundI( ebp, eax, ebx, ecx, edx, 10, 15, k3)
    ASMSubroundI( edx, ebp, eax, ebx, ecx,  0, 14, k3)
    ASMSubroundI( ecx, edx, ebp, eax, ebx,  8, 15, k3)
    ASMSubroundI( ebx, ecx, edx, ebp, eax, 12,  9, k3)
    ASMSubroundI( eax, ebx, ecx, edx, ebp,  4,  8, k3)
    ASMSubroundI( ebp, eax, ebx, ecx, edx, 13,  9, k3)
    ASMSubroundI( edx, ebp, eax, ebx, ecx,  3, 14, k3)
    ASMSubroundI( ecx, edx, ebp, eax, ebx,  7,  5, k3)
    ASMSubroundI( ebx, ecx, edx, ebp, eax, 15,  6, k3)
    ASMSubroundI( eax, ebx, ecx, edx, ebp, 14,  8, k3)
    ASMSubroundI( ebp, eax, ebx, ecx, edx,  5,  6, k3)
    ASMSubroundI( edx, ebp, eax, ebx, ecx,  6,  5, k3)
    ASMSubroundI( ecx, edx, ebp, eax, ebx,  2, 12, k3)

    // setup
    AS2(    mov   esi, ebp      )

    ASMSubroundJ( ebx, ecx, edx, ebp, eax,  4,  9, k4)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp,  0, 15, k4)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx,  5,  5, k4)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx,  9, 11, k4)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx,  7,  6, k4)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax, 12,  8, k4)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp,  2, 13, k4)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx, 10, 12, k4)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx, 14,  5, k4)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx,  1, 12, k4)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax,  3, 13, k4)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp,  8, 14, k4)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx, 11, 11, k4)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx,  6,  8, k4)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx, 15,  5, k4)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax, 13,  6, k4)

    // store a1 - e1 on stack
    AS2(    movd  esi, mm1              )   // digest_

    AS2(    mov   [esp],      eax       )
    AS2(    mov   [esp +  4], ebx       )
    AS2(    mov   [esp +  8], ecx       )
    AS2(    mov   [esp + 12], edx       )
    AS2(    mov   [esp + 16], ebp       )

    AS2(    mov   eax, [esi]            )   // a2
    AS2(    mov   ebx, [esi +  4]       )   // b2
    AS2(    mov   ecx, [esi +  8]       )   // c2
    AS2(    mov   edx, [esi + 12]       )   // d2
    AS2(    mov   ebp, [esi + 16]       )   // e2


    // setup
    AS2(    mov   esi, edx      )

    ASMSubroundJ( eax, ebx, ecx, edx, ebp,  5,  8, k5)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx, 14,  9, k5)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx,  7,  9, k5)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx,  0, 11, k5)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax,  9, 13, k5)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp,  2, 15, k5)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx, 11, 15, k5)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx,  4,  5, k5)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx, 13,  7, k5)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax,  6,  7, k5)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp, 15,  8, k5)
    ASMSubroundJ( ebp, eax, ebx, ecx, edx,  8, 11, k5)
    ASMSubroundJ( edx, ebp, eax, ebx, ecx,  1, 14, k5)
    ASMSubroundJ( ecx, edx, ebp, eax, ebx, 10, 14, k5)
    ASMSubroundJ( ebx, ecx, edx, ebp, eax,  3, 12, k5)
    ASMSubroundJ( eax, ebx, ecx, edx, ebp, 12,  6, k5)

    // setup
    AS2(    mov   esi, ebx      )

    ASMSubroundI( ebp, eax, ebx, ecx, edx,  6,  9, k6) 
    ASMSubroundI( edx, ebp, eax, ebx, ecx, 11, 13, k6)
    ASMSubroundI( ecx, edx, ebp, eax, ebx,  3, 15, k6)
    ASMSubroundI( ebx, ecx, edx, ebp, eax,  7,  7, k6)
    ASMSubroundI( eax, ebx, ecx, edx, ebp,  0, 12, k6)
    ASMSubroundI( ebp, eax, ebx, ecx, edx, 13,  8, k6)
    ASMSubroundI( edx, ebp, eax, ebx, ecx,  5,  9, k6)
    ASMSubroundI( ecx, edx, ebp, eax, ebx, 10, 11, k6)
    ASMSubroundI( ebx, ecx, edx, ebp, eax, 14,  7, k6)
    ASMSubroundI( eax, ebx, ecx, edx, ebp, 15,  7, k6)
    ASMSubroundI( ebp, eax, ebx, ecx, edx,  8, 12, k6)
    ASMSubroundI( edx, ebp, eax, ebx, ecx, 12,  7, k6)
    ASMSubroundI( ecx, edx, ebp, eax, ebx,  4,  6, k6)
    ASMSubroundI( ebx, ecx, edx, ebp, eax,  9, 15, k6)
    ASMSubroundI( eax, ebx, ecx, edx, ebp,  1, 13, k6)
    ASMSubroundI( ebp, eax, ebx, ecx, edx,  2, 11, k6)

    ASMSubroundH( edx, ebp, eax, ebx, ecx, 15,  9, k7)
    ASMSubroundH( ecx, edx, ebp, eax, ebx,  5,  7, k7)
    ASMSubroundH( ebx, ecx, edx, ebp, eax,  1, 15, k7)
    ASMSubroundH( eax, ebx, ecx, edx, ebp,  3, 11, k7)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  7,  8, k7)
    ASMSubroundH( edx, ebp, eax, ebx, ecx, 14,  6, k7)
    ASMSubroundH( ecx, edx, ebp, eax, ebx,  6,  6, k7)
    ASMSubroundH( ebx, ecx, edx, ebp, eax,  9, 14, k7)
    ASMSubroundH( eax, ebx, ecx, edx, ebp, 11, 12, k7)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  8, 13, k7)
    ASMSubroundH( edx, ebp, eax, ebx, ecx, 12,  5, k7)
    ASMSubroundH( ecx, edx, ebp, eax, ebx,  2, 14, k7)
    ASMSubroundH( ebx, ecx, edx, ebp, eax, 10, 13, k7)
    ASMSubroundH( eax, ebx, ecx, edx, ebp,  0, 13, k7)
    ASMSubroundH( ebp, eax, ebx, ecx, edx,  4,  7, k7)
    ASMSubroundH( edx, ebp, eax, ebx, ecx, 13,  5, k7)

    ASMSubroundG( ecx, edx, ebp, eax, ebx,  8, 15, k8)
    ASMSubroundG( ebx, ecx, edx, ebp, eax,  6,  5, k8)
    ASMSubroundG( eax, ebx, ecx, edx, ebp,  4,  8, k8)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  1, 11, k8)
    ASMSubroundG( edx, ebp, eax, ebx, ecx,  3, 14, k8)
    ASMSubroundG( ecx, edx, ebp, eax, ebx, 11, 14, k8)
    ASMSubroundG( ebx, ecx, edx, ebp, eax, 15,  6, k8)
    ASMSubroundG( eax, ebx, ecx, edx, ebp,  0, 14, k8)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  5,  6, k8)
    ASMSubroundG( edx, ebp, eax, ebx, ecx, 12,  9, k8)
    ASMSubroundG( ecx, edx, ebp, eax, ebx,  2, 12, k8)
    ASMSubroundG( ebx, ecx, edx, ebp, eax, 13,  9, k8)
    ASMSubroundG( eax, ebx, ecx, edx, ebp,  9, 12, k8)
    ASMSubroundG( ebp, eax, ebx, ecx, edx,  7,  5, k8)
    ASMSubroundG( edx, ebp, eax, ebx, ecx, 10, 15, k8)
    ASMSubroundG( ecx, edx, ebp, eax, ebx, 14,  8, k8)

    ASMSubroundF( ebx, ecx, edx, ebp, eax, 12,  8)
    ASMSubroundF( eax, ebx, ecx, edx, ebp, 15,  5)
    ASMSubroundF( ebp, eax, ebx, ecx, edx, 10, 12)
    ASMSubroundF( edx, ebp, eax, ebx, ecx,  4,  9)
    ASMSubroundF( ecx, edx, ebp, eax, ebx,  1, 12)
    ASMSubroundF( ebx, ecx, edx, ebp, eax,  5,  5)
    ASMSubroundF( eax, ebx, ecx, edx, ebp,  8, 14)
    ASMSubroundF( ebp, eax, ebx, ecx, edx,  7,  6)
    ASMSubroundF( edx, ebp, eax, ebx, ecx,  6,  8)
    ASMSubroundF( ecx, edx, ebp, eax, ebx,  2, 13)
    ASMSubroundF( ebx, ecx, edx, ebp, eax, 13,  6)
    ASMSubroundF( eax, ebx, ecx, edx, ebp, 14,  5)
    ASMSubroundF( ebp, eax, ebx, ecx, edx,  0, 15)
    ASMSubroundF( edx, ebp, eax, ebx, ecx,  3, 13)
    ASMSubroundF( ecx, edx, ebp, eax, ebx,  9, 11)
    ASMSubroundF( ebx, ecx, edx, ebp, eax, 11, 11)

    // advance data and store for next round
    AS2(    add   edi, 64                       )
    AS2(    movd  esi, mm1                      )   // digest_
    AS2(    movd  mm0, edi                      )   // store

    // now edi as tmp

    // c1         = digest_[1] + c1 + d2;
    AS2(    add   [esp +  8], edx               )   // + d2
    AS2(    mov   edi, [esi + 4]                )   // digest_[1]
    AS2(    add   [esp +  8], edi               )

    // digest_[1] = digest_[2] + d1 + e2;
    AS2(    mov   [esi + 4], ebp                )   // e2
    AS2(    mov   edi, [esp + 12]               )   // d1
    AS2(    add   edi, [esi + 8]                )   // digest_[2]
    AS2(    add   [esi + 4], edi                )

    // digest_[2] = digest_[3] + e1 + a2;
    AS2(    mov   [esi + 8], eax                )   // a2
    AS2(    mov   edi, [esp + 16]               )   // e1
    AS2(    add   edi, [esi + 12]               )   // digest_[3]
    AS2(    add   [esi + 8], edi                )

    // digest_[3] = digest_[4] + a1 + b2;
    AS2(    mov   [esi + 12], ebx               )   // b2
    AS2(    mov   edi, [esp]                    )   // a1
    AS2(    add   edi, [esi + 16]               )   // digest_[4]
    AS2(    add   [esi + 12], edi               )

    // digest_[4] = digest_[0] + b1 + c2;
    AS2(    mov   [esi + 16], ecx               )   // c2
    AS2(    mov   edi, [esp +  4]               )   // b1
    AS2(    add   edi, [esi]                    )   // digest_[0]
    AS2(    add   [esi + 16], edi               )

    // digest_[0] = c1;
    AS2(    mov   edi, [esp +  8]               )   // c1
    AS2(    mov   [esi], edi                    )

    // setup for loop back
    AS2(    movd  edx, mm2              )   // times
    AS2(    movd  edi, mm0              )   // data, already advanced
    AS1(    dec   edx                   )
    AS1(    jnz   loopStart             )


    EPILOG()
833 834 835
}


unknown's avatar
unknown committed
836 837 838
#endif // DO_RIPEMD_ASM


839
} // namespace TaoCrypt